The objective of this research is to test the effects of listening to noise (i.e. auditory feedback masking) on speech fluency in stroke survivors with aphasia and apraxia of speech. People with nonfluent types of aphasia frequently have apraxia of speech, which affects the motor programming of speech movements, causing distortions, slow rate, and speech disfluencies that impede the forward flow of communication. Speaking while listening to noise (e.g. auditory masking) is known to reduce disfluencies and increase speech rate in people who stutter. This method has been tested in people with aphasia, resulting in positive effects on speech production for a subset of those tested. We contend that individuals who have apraxia of speech in addition to aphasia are most likely to benefit from auditory masking, but most previous studies did not test participants for apraxia of speech. In addition, though masking is most likely to affect speech disfluencies, previous studies did not measures disfluencies. The proposed work has two specific aims. Aim 1 will determine the effect of auditory masking on speech fluency in stroke survivors with aphasia and apraxia of speech. Aim 2 will identify individual factors that predict a positive response, including presence of apraxia of speech, lesion characteristics, and type of aphasia (e.g. Broca's, Wernicke's). Voxel-based lesion analysis techniques will be used to determine sites of lesion associated with positive and negative response to auditory masking. Completion of this study will remove barriers to studying auditory masking as a technique for clinical intervention, but also as a research tool for behavioral neuroscientists probing the speech motor control system in speakers with aphasia and apraxia of speech.